Runs Qwen Code headlessly for repository analysis and coding tasks. Use when the user asks to delegate work to Qwen Code or obtain a Qwen-based coding-agent res
复制下面这句话,粘贴给 Claude Code、Codex、Cursor 等 AI 编程工具,它会读取安装说明并在你确认后完成安装。
请阅读 https://ai.atlankj.com/install/asset/gh-code-mate-qwen-code-687ca121dfa1 ,按照其中的说明把「code-mate-qwen-code」安装到你(当前 AI 工具)中。执行前先告诉我将运行的命令和写入的位置,等我确认。
查看 AI 将读取的安装说明正在读取 GitHub 原文…
内容来自 GitHub 原始文件,由原作者维护。在 GitHub 查看
command -v qwen. If it is missing, stop and ask the user to install Qwen Code in Code Mate.qwen -p "<prompt>" --output-format json
Pass the prompt as one quoted argument. Parse the JSON result and treat a structured error or nonzero exit as failure. Never start the interactive UI or OAuth flow from the agent task.
If Qwen reports missing login, API key, model, or provider configuration, stop and ask the user to configure Qwen Code in Code Mate. Never request, read, print, or copy credentials.
Use the default restricted approval behavior for analysis. Add bounded turn or tool-call limits for large tasks. Only allow modification tools when the user explicitly requests workspace changes, and never use --yolo: it grants approval without providing a sandbox.
Example: ask Qwen to identify the cause of a failing test without editing, run the command above, then summarize the parsed result within the chosen timeout.